| 基于融合技术的BP算法在发动机诊断中的应用 |
| |
| 引用本文： | 董恩国,张蕾,杨清波.基于融合技术的BP算法在发动机诊断中的应用[J].车用发动机,2003(5):49-52. |
| |
| 作者姓名： | 董恩国 张蕾 杨清波 |
| |
| 作者单位： | 1. 天津职业技术师范学院,天津,300222
2. 辽宁省高速公路管理局,辽宁,沈阳,110000 |
| |
| 基金项目： | 天津市教委高校科技发展基金项目(01-20303) |
| |
| 摘 要： | 主要融合了怠速控制阀信号、喷油驱动器信号及水温传感器信号,设计了通过提取信号特征值,应用神经网络故障识别系统诊断发动机怠速不良故障。信号特征以各自信号的特点提取,其中怠速控制阀信号为脉宽调制信号,选取脉宽及幅值等参数为特征;喷油驱动器信号也为脉宽调制信号,选取喷油时间和峰值电压等为特征;水温信号因其为慢速变化信号,取不同阶段的水温为特征。神经网络的BP算法加入动量项法和动态参数两阶段调整法进行改进。
|
| 关 键 词： | 融合技术 BP算法 发动机 故障诊断 |
| 文章编号： | 1001-2222(2003)05-0049-04 |
| 修稿时间： | 2002年11月13日 |
The Application on BP Algorithm with Fusion Technology in Engine Fault Diagnosis |
| |
| Abstract: | In fault diagnosis system of neural net , applying fuse technology to analyze synthetically the different kinds of related signals of engine, can rapidly and accurately diagnose engine faults. The paper illustrates the diagnosis way by fusing IACV (idle air control valve)signal , SFl(sequential fuel injection) signal and ECT (engine coolant temperature) signal to diagnose idle system faults. The diagnosis is finished by choosing characters of three kinds of signals in the system of neural net. On the characters, IACV is pulse width modulating signal, which chooses pulse width and peak value, etc; fuel injection signal is also pulse width modulating signal, which chooses injection time and peak value, etc; water temperature signal chooses the temperature of different phase because of its slow variation. BP of neural net is improved by adding momentum term and adjusting dynamic parameters twice. |
| |
| Keywords: | fusion technology BP algorithm engine fault diagnosis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
